microsoft/onnxruntime

Pull Requests Commits

Try increasing android emulator memory to 6GiB

adrianlizarraga committed 21 days ago

63826c01

Fix BART attention fusion for SDPA pattern from transformers >= 4.49 (#27458)

Rishi-Dave committed 21 days ago

Verified 028f88cf

Fix Cast node naming collisions and opset 10 Resize in float16 conversion (#27469)

Rishi-Dave committed 21 days ago

Verified dce58e87

Build Windows ARM64X binaries as part of packaging pipeline (#27316)

edgchen1 committed 22 days ago

Verified ffbbc4f7

softplus support for webgpu (#27457)

guschmue committed 22 days ago

Verified f73b1d76

Remove unused cusparse library dependency (#27467)

Copilot committed 22 days ago

Verified 61244135

upgrade dawn to ec7b457e5bb1fcec6f59733c4f3dd84d2f885a38 (#27427)

fs-eire committed 22 days ago

Verified e1659122

[web] remove the unhelpful "Unknown CPU vendor" warning. (#27399)

fs-eire committed 22 days ago

Verified 1131a86a

[web] fix usage of wasmBinary together with a blob URL for .mjs (#27411)

fs-eire committed 22 days ago

Verified cf96123f

[MLAS] Add kleidiai pad ptr invalidation test case (#27465)

JonathanC-ARM committed 22 days ago

Verified d1d596c8

Avoid repetitive creation of fp4/fp8 native-custom-op domains for NvTensorRtRtx EP (#27192)

vishalpandya1990 committed 22 days ago

Verified d5387d81

Support TensorScatter (24) - CUDA (#27446)

titaiwangms committed 22 days ago

Verified 39bdea44

Add optional dependency for quantization utilities (ml_dtypes) (#27435)

chaya2350 committed 22 days ago

Verified 7c56ee85

Fix 3d attention mask broadcasting in MHA (#27464)

titaiwangms committed 22 days ago

Verified 654c335c

Fix GatherCopyData Integer Truncation Leading to Heap Out-of-Bounds Read/Write (#27444)

chilo-ms committed 22 days ago

Verified 355278a3

Update python_requires (#27354)

XXXXRT666 committed 22 days ago

Verified eb23be8d

Fix SkipLayerNorm fusion incorrectly applied when gamma/beta are not 1D (#27459)

Copilot committed 22 days ago

Verified 82eb99c0

optimized qmoe code path for 1 token (#27383)

guschmue committed 23 days ago

Verified 5f087c41

[wasm] add flags for WebGPU build to reduce binary size (#27371)

fs-eire committed 23 days ago

Verified 3db53eb0

WebGPU: Optimize GraphCacheManager by removing pending_buffers (#27453)

qjia7 committed 23 days ago

Verified c6491e0e

[MLAS] ukernel interface logging changes (#27439)

JonathanC-ARM committed 23 days ago

Verified 98cc06e8

Fix VersionNumber usage in React Native Android build with major version validation (#27385)

Copilot committed 23 days ago

Verified f3a62845

Add OrtModel input support for Compile API (#27332)

adrastogi committed 23 days ago

Verified 500ab5f1

fix: OrtValue.from_dlpack rejects zero-size tensors as non-contiguous (#27451)

Copilot committed 23 days ago

Verified 616a4dfc

webgpu: support head_sink in flash attention (#27410)

guschmue committed 23 days ago

Verified bb3866cf

Improve Pre Packing for 2 bit LUT kernels (#27131)

vraspar committed 23 days ago

Verified 2145c8c0

fix: out of bounds access for resize operation (#27419)

lukas-folle-snkeos committed 24 days ago

Verified 79e0676e

Add validity check on MaxUnpool indicies. (#27432)

skottmckay committed 24 days ago

Verified 1f05a4ce

webgpu: Add configurable multi rotary cache concat offset parameter (#27434)

qjia7 committed 24 days ago

Verified 88855c7f

Enable Python 3.14 CI and Upgrade Dependencies (#27401)

tianleiwu committed 24 days ago

Verified 71d2cd01

Older